Human cytomegalovirus and Epstein-Barr virus infections, risk factors, and their influence on the liver function of patients with acute-on-chronic liver failure

Human cytomegalovirus (HCMV) is a β human herpesvirus with positive rates of antibodies of 50–100% [1, 2]. Primary HCMV infection is often acquired during early childhood, and in later development stages, HCMV establishes lifelong latency or persistence within a person through cells of myeloid lineage [1] or granulocyte-monocyte lineage [3]. However, reactivated or exogenous reinfection with HCMV may occur and may become a major viral cause of morbidity and mortality in immunocompromised patients, such as organ or haematopoietic stem cell transplant (HSCT) recipients [4‐6] and patients with acquired immune deficiency syndrome (AIDS) [1, 2].

Epstein-Barr virus (EBV) is a γ human herpesvirus with positive rates of antibodies of up to 90% [7, 8]. Primary EBV infection mostly manifests as infectious mononucleosis, with fever, angina, lymphadenopathy, and liver and spleen enlargement. After a primary infection occurs, EBV also develops a lifelong latency in B cells [9]. Latent EBV may be reactivated under certain conditions, thereby inducing abnormal B cell proliferation and leading to neoplastic disease.

Studies on HCMV have focused mostly on immunosuppressed individuals [2, 4, 5]. Studies on EBV have mostly examined patients with neoplastic disease [10] and post-transplant lymphoproliferative disorder following HSCT [5, 11, 12]. Limited studies have considered immunocompetent and not severely immunocompromised patients.

Acute liver failure (ALF) is a well-established medical emergency defined as a severe liver injury. However, a proportion of patients who present features mimicking ALF suffer from an underlying chronic liver disease or liver cirrhosis. These patients, who have been defined as patients with acute-on-chronic liver failure (ACLF), display various degrees of immune disorders and short-term mortality rates [13]. HCMV may induce fulminant hepatic failure [14, 15], and EBV may trigger acute liver failure [15‐17].

HCMV may be reactivated in patients with cirrhosis [18‐20]. Rosi, S. et al. [19] recently revealed that HCMV causes or contributes to hepatic decompensation or ACLF in patients with cirrhosis even if they are not severely immunocompromised. However, the patients in the above study [19] had alcohol-related and mixed-aetiology cirrhosis (hepatitis C virus and alcohol-related cirrhosis). To date, studies have yet to describe HCMV or EBV infections in patients with ACLF due to hepatitis B virus (HBV)-related chronic liver disease or liver cirrhosis.

Therefore, this study aims to identify HCMV and EBV infection in patients with ACLF-based HBV in our hospital. The specific objective of this study was to explore the HCMV and EBV infection rate, the risk factors of these infections, and their influence on the liver function of patients with ACLF-based HBV, which will improve the survival of these patients through an effective intervention.

Human herpes viruses include herpes simplex virus type 1 (HSV-1), herpes simplex virus type 2 (HSV-2), varicella zoster virus (VZV), HCMV, EBV, human herpesvirus 6 (HHV-6), human herpesvirus 7 (HHV-7), and human herpesvirus 8 (HHV-8). Although HCMV and EBV are more studied in this respect, after initial infections with no or mild symptoms, all human herpes viruses may become latent under certain conditions. However, most studies of latent HCMV and EBV infection have focused on immunosuppressed individuals. In this study, we aimed to focus our investigation on HCMV and EBV in immunocompetent individuals.

Our study evaluates a cohort of 100 hospitalized patients with ACLF due to HBV infection, of which 5.0% patients presented with HCMV DNA, 23.0% patients presented with EBV DNA, and one patient was positive for both HCMV DNA and EBV DNA. HBV DNA < 1000 IU/mL was a risk factor for HCMV infection, and age, especially older than 60 years, were risk factors for EBV infection. HCMV-infected patients had lower ALT levels, indicating that HCMV infection may not increase liver injury. In contrast, EBV infection possibly influenced liver function, as EBV-infected patients had lower albumin levels and higher Child-Pugh scores.

By reviewing the medical records and laboratory data, we found that of the 69 patients in this study in whom HCMV IgG and EB viral capsid antigen (EB-VCA) IgG were measured, all were HCMV IgG- and EB-VCA IgG-positive. HCMV and EBV primary infection mostly occur in childhood. Thus, in this study, it is likely that the observed HCMV and EBV infections were due to reactivation rather than primary infection.

Previous studies have investigated the association between HCMV infection and HBV infection [23, 28‐30]. In the present study, we found that 5.0% of patients were co-infected with HCMV, and this finding was similar to that reported by Lian et al. [30]. Limited studies have explored the association between EBV infection and HBV infection [25]. An et al. [25] reported that compared with HBV infection (9.10%), EBV infection occurred more frequently in patients with HBV-related liver cirrhosis (40.0%) and liver carcinoma patients (25.0%). In the present study, we found that 23.0% of the patients were infected with EBV, which differed from the previously reported infection rate because chronic HBV infection and liver cirrhosis were both included in our study.

Immunosuppressive drugs, high-dose corticosteroids, T-cell depletion, acute and chronic GVHD, rejection, and virus coinfection were identified as the most common risk factors for HCMV or EBV infection [3, 5, 6, 11, 12, 22]. These risk factors were determined mainly based on immunodeficient or immunosuppressed patients, such as organ transplant recipients, HSCT recipients, and patients with AIDS. Few studies have considered immunocompetent and not severely immunocompromised patients. In the present study, we focused on patients with ACLF due to HBV infection.

ACLF is defined as an acute hepatic insult manifesting as jaundice and coagulopathy that is complicated within 4 weeks by ascites and/or encephalopathy in patients with previously diagnosed or undiagnosed chronic liver disease [13, 21] with various degrees of immune disorders. Considering the relationship among HCMV, EBV, and HBV [23, 25, 28‐30], we attempted to explore whether HBV-related indexes and liver cirrhosis are potential risk factors in addition to age and gender.

We found that HCMV-infected patients had lower HBV DNA levels than HCMV-uninfected patients (P = 0.001). Eighty percent of the HCMV-infected patients had HBV DNA < 1000 IU/mL, which was greater than the number of HCMV-uninfected patients with similar levels (9.5%). In addition, we found that HBV DNA < 1000 IU/mL was a risk factor for HCMV infection. HBV was strongly correlated with HCMV. Our findings were similar to those of previous reports [23, 24]. Cavanaugh, V. J. et al. [24] discovered that cytokines produced by murine cytomegalovirus inhibited HBV replication and gene expression. Bayram, A. et al. [23] revealed that the inflammation caused by HCMV can contribute to viral clearance during chronic HBV infection. Therefore, we suggest that HBV and HCMV replication may mutually inhibit each other. Thus, when HBV replication is inactive, as indicated by a lower titre of HBV DNA, especially HBV DNA < 1000 IU/mL, HCMV replication activity is facilitated. However, the specific mechanism underlying this phenomenon must be further studied.

Our previous study posited that male donors, conditioning regimens including ATG and GVHD prophylaxis, and prednisone increase the risk of EBV infection in patients receiving HSCT [5]. However, EBV infection and its risk factors were rare in patients who are not severely immunocompromised, such as those with ACLF. In the present study, HBV-related indexes, including proliferation state, serological status of HBV and liver cirrhosis, were not risk factors of EBV infection in HBV-related ACLF. However, age was a risk factor for EBV infection, especially in patients > 60 years of age. XM. Zhang [31] previously reported that the rate of EBV-1 infection in this age group was significantly higher than that in the < 40 age group. Age was related to EBV-1 infection in patients with chronic periodontitis. It was suggested that the immune system and the environment of the host change with the age, and thus, latent EBV was released and propagated in the host cell [31].

Bayram, A. et al. [23] indicated that the mean ALT and intrahepatic HBV DNA levels in patients with HCMV co-infected with chronic viral hepatitis B were lower than those in patients without HCMV co-infection. Similarly, we found that the levels of ALT, AST, and HBV DNA were lower in patients with ACLF due to HBV co-infection with HCMV. However, further analysis using binary logistic regression indicated that only the ALT level was related to HCMV infection. Therefore, we suggest that HCMV infection may inhibit HBV proliferation, decrease inflammatory activity, reduce liver cell damage, and reduce ALT levels. However, the specific mechanism underlying this phenomenon must be further studied.

Previous studies have reported that EBV infection causes acute liver failure [15‐17]. In the present study, EBV-infected patients had lower albumin levels than uninfected patients. In addition, EBV-infected patients also had higher Child-Pugh scores. Thus, EBV infection may be related to lower albumin and higher Child-Pugh score. EBV infection may increase liver injury and decrease the liver synthesis ability in ACLF patients. Furthermore, Child-Pugh score is a suitable prognostic evaluation indictor for patients with end-stage liver disease; a higher score corresponds to higher mortality in the short term [26]. Thus, the influence of EBV infection on liver function may result in poor prognosis.

This work is limited by a number of factors. First, a slight bias is present in any retrospective study, which will result in a certain statistical bias. Second, we did not conduct follow-up for the included patients and did not analyse the confirmed influence of prognosis on HCMV and EBV infection in the short and long terms. Third, we unfortunately did not include all human herpes viruses in this study. Fourth, we found that HCMV-infected patients had lower HBV DNA levels, and thus HCMV viral replication is likely to be detrimental to HBV replication in these 5 HCMV-infected cases. This may be due to the inflammatory cytokines caused by HCMV contributing to viral clearance during chronic HBV infection. However, we unfortunately did not measure the proinflammatory cytokines released by HCMV that prevent HBV replication. Fifth, we cannot distinguish between primary infection and reactivation of HCMV and EBV infection for all patients, although there all 69 patients in our study who were evaluated were found to be HCMV IgG- and EBV IgG-positive. We will address these drawbacks in our future studies. Finally, although we found that EBV-infected patients had lower albumin levels and higher Child-Pugh scores, as indicated by binary logistic regression, this may be associated with overall physiological decline in these patients. Although this study does have limitations, our evaluation of patients with HBV-related ACLF may begin to fill in gaps of knowledge on HCMV and EBV infection in immunocompetent patients.

This study revealed that HCMV and, more notably, Epstein-Barr virus (EBV) were detected in patients with ACLF due to HBV infection. Lower replication levels of HBV (HBV DNA < 1000 IU/mL) was associated with an increased probability of HCMV infection; age, especially older than 60 years, were associated with EBV infection. HCMV infection may inhibit HBV proliferation and reduce liver injury, while EBV infection possibly influenced liver function, as EBV-infected patients had lower albumin levels and higher Child-Pugh scores. HCMV co-infection did not increase liver injury. However, co-infection with EBV may influence liver function and may induce poor prognosis in HBV-related ACLF. Thus, attention should be paid to the possibility that older patients, especially those older than 60 years with ACLF due to HBV, may be infected with EBV. In this case, effective antiviral therapy may be needed; otherwise, EBV infection may influence the patients^,prognosis.